Qlik Connect 2026 | Drew Clarke & Juan Hurtado, Qlik

Clips
More from Qlik Connect 2026

Drew Clarke

EVP, Product & Technology

Qlik

Juan Hurtado

VP, BI & Data Analytics, Ingersoll Rand & Executive Advisory Board, Qlik

Ingersoll Rand

play_circle_outline Qlik as AI Agent: Cut 70–80% Data Engineering Overhead by Staying Close to Data Gravity and BI for Mixed Data

play_circle_outline Conversational Analytics and Decision Automation: AI as Business Partner Unlocking Institutional Knowledge

play_circle_outline "AI factory" concept: standardized, composable architecture respecting semantic layers

play_circle_outline Multi-Cloud Lakehouse Interoperability: Qlik Native Iceberg, Medallion Architecture and the Goldilocks Data Zone

Info
Transcript

Drew Clarke & Juan Hurtado, Qlik

Drew Clarke

EVP, Product & Technology Qlik

Juan Hurtado

VP, BI & Data Analytics, Ingersoll Rand & Executive Advisory Board, Qlik Ingersoll Rand

In this interview from Qlik Connect 2026, Drew Clarke, executive vice president of products and technology at Qlik Technologies Inc., joins Juan Hurtado, vice president of business intelligence and data analytics at Ingersoll Rand Inc. and executive advisory board member at Qlik, to talk with theCUBE Research's Rob Strechay about how enterprises are moving beyond dashboards to AI-powered decision automation. Clarke and Hurtado explore why 70-80% of enterprise AI efforts stall in data engineering rather than delivering business value — and how anchoring AI ini... Read more

explore Keep Exploring

How did the advisory board's advice to focus Qlik as the AI agent for analytics shape your strategy, and what lessons have you learned building enterprise-grade AI solutions? add

Are organizations moving beyond dashboards toward AI-powered, insight-led decision automation that turns data into actionable insights? add

What are the key design principles for building a repeatable, composable "AI factory" architecture that leverages lakehouse data gravity and semantic layers while minimizing technical debt? add

What are the benefits of adopting a lakehouse architecture (with Iceberg support) for organizations that have data spread across multiple platforms, and how does it help reduce data replication, improve interoperability, and simplify maintenance? add

bolt Powered by CUBE AI

Drew Clarke & Juan Hurtado, Qlik

search

Rob Strechay

>> Hello and welcome back to Qlik Connect 2026 from lovely Orlando, Florida. I'm Rob Strechay, and I'm really excited to be here. The people who know me, this is near and dear to my heart where data and AI meet and really produce ROI. And I get to talk about that with Drew Clarke, who's the EVP of product and technology for Qlik, and Juan Hurtado.

Juan Hurtado

>> Yeah, that's right.

Rob Strechay

>> I got it right.

Juan Hurtado

>> There you go.

Rob Strechay

>> That's the important part of my job. Juan Hurtado, who's the VP of BI and data analytics at Ingersoll Rand and an executive advisory board for Qlik. So welcome on board, both of you.

Drew Clarke

>> Yeah, glad to be here.

Juan Hurtado

>> Thank you for having us.

Rob Strechay

>> Yeah, so fresh off the main stage, this has been a lot of fun, there's a lot of energy going on here, a lot of customers here. What have you seen that has really helped people ensure that they're getting their insights seen and that they're seeing the product help them move from pilot to production, because that's what I always get asked is, and data quality and data fragmentation is a big thing around that? And you had a lot of things on the main stage.

Drew Clarke

>> I did, and actually, I'd love to bring back to conversation one, you and I had a little over a year ago, and we were talking about AI and the role of generative AI and being able to improve productivity enhancements. And you were very explicit to us to say, "Look, Qlik, I need you to focus and be the AI agent for the analytics. Don't try to be in all the different aspects, you have a role to play." And helped us focus to say, all right, we're going to orient on the business analyst. We're going to orient on the data engineer, building the tools into your factory that you talked about today. But you gave us that advice as part of the advisory board, and we took it to heart. I think we did.

Juan Hurtado

>> That's right. For us has been like a journey. We have failed and we have learned some hard lessons. Lessons like, for example, to make great enterprise grade AI, you probably need a good mix of unstructured and structured data. The first time we approached this was like, we don't know exactly what we're trying to do. We go and pay big money to some consultancy companies to help us build these solutions. What happens is that we see 70% or 80% of that effort being data engineering on the structured data and unstructured data. Then 20% is that really myth of building the AI solution. And then what happens is one month after the solution has been delivered, the application is already obsolete and deprecated. And we are back to data engineering and the data needs to be updated and so on. So when we started to think, "Hey, how do we get out of this value of ," because everyone is talking about like 95% of all the enterprise grade AI failing. And we said, "Look, one thing for sure is stay close to the data gravity of your data and BI platforms." Qlik was there for us. We started to look at how do we make conversational analytics and good agents out of the data that we already curated, and we didn't need all these data engineering efforts.

Drew Clarke

>> Yeah. Yeah, so not starting from scratch, but starting ahead with already where you had that data gravity.

Juan Hurtado

>> That's right.

Drew Clarke

>> Right.

Rob Strechay

>> To me, that seems like how do you get past dashboards and get to really decision automation. And you were talking about it on stage earlier today, it's not about just having yet another dashboard, although it can help do that to represent data, but it's more talking with the data and bringing it to life.

Juan Hurtado

>> We like to say that our users are quickly evolving. They will very soon become not only dashboard consumers, but what they're looking is really to have like an AI powered solution that can really become their own analytics business partner. And that's what you achieve when you combine the data platform, the data gravity of the business intelligence solution, and the AI capabilities.

Rob Strechay

>> Yeah. That's to me, it was the example you were showing on stage around fall risk and things like that was very much, it's not about just having a dashboard, it's actionable insights into this that then are powered and keep going on. Are you seeing that organizations are really looking for insight led innovation from that perspective?

Drew Clarke

>> Well, what they're showing, and you said this, Juan, earlier, was you have 1,500 professionals who are using data to do their job to be able to get to the outcome. And what you've done is powered them or given them the AI to do their job more efficiently and effectively. And I thought that just captured really well of it's beyond dashboards. Actually, the logic, and you talked about the business logic that goes into the curation, the management, the pipelines that feed it, and now opening it up into that new environment, really gets you to the actions, the decisions. And that's where value comes.

Juan Hurtado

>> And I like to think that the people that we have for a long time trained that are now knowledgeable about our data, the domain knowledge, understand patterns, are outperforming the specialists on the field. And so with the help of AI, of course, so we are unlocking this power of our institutional knowledge that leaves with the data practitioners, and now they're using that with a lot of creativity and innovation and applying AI, and now giving a speed and velocity to decision making and so on.

Rob Strechay

>> So do you have key design principles that help you steer people so that they really make it repeatable, and so that they can really lean into this as they go, as you've learned?

Juan Hurtado

>> We call it the AI factory. So it is a principle of like, we want a, let's say, somehow a standardized composable architecture where parts of that architecture lies on the fact that we respect the data gravity of the lakehouse, we respect the semantic layers that have been built painstakingly within Qlik. We respect some basic principles of how we treat the unstructured data and the structured data, because that's all what is going to accelerate the speed of delivery of the solutions without making every single solution a point to point solution, or something where it's going to create technical debt very quickly.

Rob Strechay

>> Yeah. Actually, you hit on a key point that came up on the main stage this morning as well around the lakehouse. And the fact that organizations, when we look at it, they have maybe 50% of their data in SQL databases, maybe 20% in Snowflake, another 20% in a Databricks, and then here and there and all over the place kind of thing. When you look at the lakehouse architecture and Qlik is leaning into iceberg, which I think is perfect, and we also see that that definitely has won, I think, and is that really one of the things, like you said, so you're not replicating, you don't have the cost of replicating that data all over the place? Is that one of the things that's a key too?

Juan Hurtado

>> It is. For example, in Ingersoll Rand, we are multi-cloud in principle. Now, in reality, it's like we have some of the data like IOT in one platform, some of the transactional in another. The way we are creating interoperabilities, for example, with Iceberg tables between these platforms, trying to work on the building edge of these technologies. And now that Qlik is like fully native support to Iceberg tables, that's going to help us accelerate a lot with our development.

Drew Clarke

>> You talked about that, maybe this is where you're going around, there's a medallion architecture, bronze to get into the silver, gold, and then the diamond. And you're going to have to trademark this, the Goldilocks zone, the Goldilocks zone where you're combining everything together. But that's where we're seeing a lot of customers, it's, all right, you put that architecture for scale, but you have to recognize that data sits in different places. And we're set up to be interoperable. We want to be interoperable. Don't create a monolith of everything. Have the data where it needs to be, make it accessible, bring it back just right, like Goldilocks.

Juan Hurtado

>> That's right. And try to avoid the dead ends from a data standpoint, interoperability, design , that's where we are seeing that the repeatability and the maintenance of the solution is sitting on that, let's say Goldilocks zone of... Yeah.

Rob Strechay

>> I would assume that that also helps when you go to other parts of Ingersoll Rand with this that may have not been the first ones in on this and are still figuring out their use cases or their processes that they want to bring into AI, but that they don't have to go retool everything and go and consolidate that.

Juan Hurtado

>> That's right, yeah. Those architectural patterns, and let's say AI factory, help us do something that has been preached for a long time. You do POCs quickly, you prototype quick, and then you scale to enterprise grade. You scale small. Typically, within a single business unit and a single geography, when things are working, with people feel that a solution can be trusted, then you quickly use the same pattern, the same architecture to expand to other business units.

Rob Strechay

>> So when you look at this, again, being multi-cloud by design or on purpose, when you start to look at governance, now, a lot of people think of governance as a bad word because it slows you down and stuff like that, but it's also, we were just talking with James about this, about the fact that you got to go slow to go fast sometimes and having those things in place. How do you see and how does Qlik help you really have that governance and control in that really multi-cloud environment that may have a lot of different people that are not within your span of control from that perspective?

Juan Hurtado

>> Yeah. It's a great question. And I think governance was a word that we, for long time, struggle a lot to communicate at executive level, because basically, it's almost like saying, "Hey, we want to be the data cops." And you don't want to be the data cop. But AI has facilitated the discussion. Now, it's very clear that you need great quality, great governance to make AI work the way it's supposed to be and trusted. And here is, with an architecture that is multi-cloud in nature, where we have a medallion architecture that span across different systems, is where you could rely on tools that are interoperable, are powered by AI because there's like this almost like a virtuous circle of like, you have governance, you let AI assess your data quality and governance, AI help you get better, and it almost like starts to create its own self-healing and a self-improving solution for governance.

Drew Clarke

>> Well, I was thinking about as you talk about that governance and then you layer in, you said you had 70 ERPs, but you do how many acquisitions a year?

Juan Hurtado

>> That's a consequence of acquisitions. We do an acquisition on once a month.

Drew Clarke

>> Once a month?

Juan Hurtado

>> Yeah.

Drew Clarke

>> Now, you think about the data architecture and what you're doing, creating your factory, the governance, and every month, you have a new company coming in that you have to integrate and bring that data together.

Juan Hurtado

>> With its own cultural and historical background, the way that metrics were interpreted differently. And so we got to build this almost serialized process to bring integrations into our data state, yeah.

Drew Clarke

>> Yeah.

Rob Strechay

>> Yeah, I would assume that it helps that you can bring that semantic knowledge into the tool as well because when you have the meaning for the data and that becomes standardized, so revenue means revenue, every acquisition now gets standardized on that semantic layer, and you're able to then... Because public company, you have a lot of reporting and all of this fun jazz that the SEC regulates and stuff.

Juan Hurtado

>> From master data to the way that every acquisition or every business unit sees the world, different geography hierarchies, different leadership reporting structures, different product lines and the way that product hierarchies are, it's something that you contextualize thanks to solutions like Qlik or powered by AI, and make sure that at the end, everything is getting aggregated and makes sense for global solutions as well as localized solutions.

Rob Strechay

>> So as we come to the end of this, my last question to both of you is, when we're sitting here a year from now, what do you hope to say that you can't say today that you want to say by next year?

Drew Clarke

>> So the velocity from event to outcome compresses, so speeds up, and that companies are able to get competitive advantage using our platform, of course, because of that velocity and that loop of event to decision to action gets faster and faster and powered by our solutions.

Juan Hurtado

>> It's very similar. I think as a segue to what Drew mentioned is we would love to see our entire data champions community that today may feel behind on like, "I'm in the AI wagon or not? I've been like a traditional AI or data citizen," we would like to see them empowered, like feeling like an AI experts, being able to develop at the speed of AI experts, thanks to the commoditization of these technologies that platforms like Qlik brings to the table. We want to see everyone empowered and like an an army of AI people that we just need to upskill a little bit and get them with the platform.

Drew Clarke

>> All right, so that was pretty cool. I did the technical answer and you did the people side. So inspiring. I love that.

Rob Strechay

>> I love both of those answers. I think again, I'm hoping for that because then I want an update on that because I think a big piece of it is bringing everybody there and leveling the playing field. And I think AI can do that, is to help to level the playing field for a lot of different people. And I'm looking forward to that too. And I think you need a strong platform to do it. And I love what you're doing in the platform.

Drew Clarke

>> Oh, thank you.

Rob Strechay

>> So Drew, Juan, thanks for coming on board.

Juan Hurtado

>> Thank you so much for inviting us.

Rob Strechay

>> And we'll see you back here next time.

Juan Hurtado

>> Good deal.

Drew Clarke

>> Thank you.

Juan Hurtado

>> Thank you.

Rob Strechay

>> But don't you go away because we have more to come from Qlik Connect 2026 live from Orlando, we'll be back shortly.

Juan Hurtado

>> Thank you.